Skip to main content

​
Overview

ChatAds supports the Model Context Protocol (MCP) for seamless integration with AI assistants like Claude, GPT, and others.

​
MCP Wrapper SDK

Install the ChatAds MCP wrapper: 
npm install chatads-mcp-wrapper

​
Configuration

Add ChatAds to your MCP server configuration: 
{
  "mcpServers": {
    "chatads": {
      "command": "npx",
      "args": ["chatads-mcp-wrapper"],
      "env": {
        "CHATADS_API_KEY": "cak_your_api_key"
      }
    }
  }
}

​
Available Tools

​
chatads_analyze

Analyze a message for affiliate product recommendations. Parameters:
  • message (required): The user message to analyze
  • country (optional): ISO country code
  • language (optional): ISO language code
Example Usage in AI Assistant: 
User: Can you find me affiliate links for yoga mats?
Assistant: [Uses chatads_analyze tool]

​
Integration Examples

​
Claude Desktop

Add to your Claude Desktop config (~/.config/claude/mcp.json): 
{
  "mcpServers": {
    "chatads": {
      "command": "npx",
      "args": ["chatads-mcp-wrapper"],
      "env": {
        "CHATADS_API_KEY": "cak_your_api_key"
      }
    }
  }
}

​
Custom MCP Server

import { Server } from "@modelcontextprotocol/sdk/server/index.js";
import { ChatAdsClient } from "@chat-ads/chatads-sdk";

const client = new ChatAdsClient({
  apiKey: process.env.CHATADS_API_KEY
});

const server = new Server({
  name: "chatads",
  version: "1.0.0"
});

server.setRequestHandler("tools/call", async (request) => {
  if (request.params.name === "chatads_analyze") {
    const { message, country, language } = request.params.arguments;

    const response = await client.analyzeMessage({
      message,
      country,
      language
    });

    return {
      content: [{
        type: "text",
        text: JSON.stringify(response, null, 2)
      }]
    };
  }
});

​
Response Format

The MCP tool returns the standard ChatAds response: 
{
  "success": true,
  "data": {
    "matched": true,
    "ad": {
      "product": "Yoga Mat",
      "link": "https://amazon.com/...",
      "message": "Perfect for beginners",
      "category": "fitness"
    }
  }
}

​
Best Practices

Use in server context

MCP servers run server-side, keeping your API key secure.

Handle errors gracefully

Implement proper error handling for rate limits and network issues.

Cache when possible

Consider caching responses for repeated queries to save API calls.

Set appropriate timeouts

Configure reasonable timeouts for the MCP tool calls.